Alfalfa (Medicago sativa L.) crop vigor and yield characterization using high-resolution aerial multispectral and thermal infrared imaging technique

Alfalfa (Medicago sativa L.) is an important forage crop grown worldwide for animal feed, green manure, and as a land cover. However, very few approaches exist timely field scale mapping of status, yield quality attributes management inputs, harvest storage resources, budgeting, insurance, etc. This study aims to apply high-resolution aerial multispectral thermal infrared remote sensing (7 cm/pixel) characterize above attributers. Imaged were two cutting cycles in 2018 season. Eight vigor index (VI) Crop Water Stress Index (CWSI) features derived from collected imagery data. Modified Non-Linear (MNLI), Simple Ratio (MSR) CWSI reliably evaluated the spatial variations stress traits (Coefficient variation [CV] ranges 24–69%). Yield was then predicted with indices predictor variables through nine simple linear regression (LRs, variable: one image feature per model), seven multiple (MLRs, variables: VI stepwise (SLR), partial least square (PLSR) absolute shrinkage selection operator (LASSO) models. The SLR, PLSR LASSO initially used all model training. Amongst models, MLR-4 (Variables: MNLI CWSI) performed best (Root mean error [RMSE] = 0.45 kg, R2 0.64) LR-5 (Variable: MNLI) second-best (RMSE 0.51 0.54). complex models similar accuracy 0.45–0.46 0.63–0.64). (canopy vigor) (stress) significant sufficient effective alfalfa status prediction their non-saturation non-linearity features. Overall, visible-NIR domain showed potential site-specific monitoring.